Key result
Physiology-aware graph attention classifier predicts sustained ventricular reentry with ~95% accuracy in simulations.
Why the study?
Post-infarct ventricular tachycardia reentry risk is difficult to predict from infarct imaging alone, and exploring many substrates and pacing locations with in silico induction studies remains computationally expensive.
Does a physiology-aware graph attention classifier accurately predict sustained ventricular reentry from early activation features in simulated post-infarct cardiac electrophysiology meshes compared to baseline machine learning models and full simulations?
Does a physiology-aware graph attention classifier accurately predict sustained ventricular reentry from early activation features in simulated post-infarct cardiac electrophysiology meshes compared to baseline machine learning models and full simulations?
A physiology-aware graph attention network can accurately predict sustained post-infarct ventricular reentry from just the first two paced activations in silico, screening candidate substrates nearly five orders of magnitude faster than full electrophysiology simulations.
Answers are grounded in this paper and cite their sources.
May accelerate in silico screening of reentry substrates; leaves open clinical translation pending in vivo validation.
Sudden cardiac death mostly originates from post-infarct ventricular tachycardia. This reentry is sustained by propagation around scar tissue, but its risk remains difficult to predict from infarct imaging alone. While in silico induction studies can reveal these mechanisms, exploring many substrates and pacing locations remains computationally expensive. We propose a graph attention classifier for early prediction of sustained ventricular reentry from simulated cardiac electrophysiology. Each simulation is represented as a graph with early nodal voltage samples and edge attributes encoding geometry, stimulation frequency, tissue labels, amplitude, and diastolic interval. These descriptors are computed only from the first two paced activations, before sustained reentry is observed. On a balanced independent test set of 1000 simulations, the classifier achieves 94.6% accuracy, 95.4% sensitivity, 93.8% specificity, 94.6% F1-score, 0.987 ROC-AUC, and 0.985 PR-AUC. Ablation experiments show that diastolic interval is the most critical descriptor. Qualitative attention inspection suggests that different heads focus on border-zone, stimulation location, and early propagation path. These results support graph attention as a fast early-prediction approach for large-scale in silico ventricular reentry studies.
No takes yet. Share an insight, caveat, or question.
Morier et al. (2026) studied Post-infarct sustained ventricular tachycardia (in silico cardiac electrophysiology simulations) (n=8,004). Physiology-aware residual Graph Attention Network (GATv2) classifier vs. Full Mitchell-Schaeffer electrophysiology simulation pipeline and architectural/non-graph baselines (Sequential GAT, GCN, SVM-RBF, NodeMLP) was evaluated on Early prediction of sustained ventricular reentry (persisting until t = 6 s) on the balanced independent test set (n=1,000). The physiology-aware graph attention classifier predicted sustained ventricular reentry with 94.6% accuracy, 95.4% sensitivity, 93.8% specificity, and 0.987 ROC-AUC on 1,000 test simulations.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: